MS6 Suspension Data & DIY Bilstein Coilovers

[phate]

Originally Posted by Nliiitend1 I wouldn't worry too much. That new spring is going to compress a LOT less than the old one at static height... Are you gonna use simple flat Torrington/needle bearings on the perches in the front?

Yeah, I'm going to wait and see about the ride height. I'm not real concerned about it.

Eibach, Hypercoil, and Rogue Engineering all said don't worry about bearings with these short springs. So, no plans for them right now.

[zenit]

An alternative to the bearing is a single sheet of Teflon or swift's spring sheets. Cheap, thin and not prone to contamination.

Sent from my SAMSUNG-SGH-I847 using Tapatalk 2

[Nliiitend1]

Originally Posted by phate Yeah, I'm going to wait and see about the ride height. I'm not real concerned about it. Eibach, Hypercoil, and Rogue Engineering all said don't worry about bearings with these short springs. So, no plans for them right now.

Yeah I have to keep reminding myself that these aren't MacPhailsons... Your steering axis is completely different than the damper axis.

[phate]

Had some time to test fit the rear spring and adjuster setup, today. This is going to work perfectly. The only thing that sucks is the adjuster is a huge pita to adjust if there's any sort of pressure on it. When it's at the highest setting, you have to drop the LCA in order to relieve pressure. This is what it looks like:



Ride height at the tallest:




Ride height at lowest setting:



^This is with the jam nut and rubber noise damper installed on the adjuster. You could go lower than this if you really wanted to, by maybe another .25-.5". If you're in this thread for the lowz, though; you're in the wrong place

Adjustment range (fender to hub center measurements):
Tallest - 15.68" (approximately .5" above stock)
Lowest - 13.625" (approximately 1.5" below stock)
I will likely end up ~1" below stock. That leaves ~3" of bump travel.


I drove it at both settings and it's way more tame than I expected. Albeit, I'm on stock shocks so it's way underdamped, but the ride isn't harsh at all. It'll probably firm up some with the revalved shocks, so I'm really looking forward to seeing what the whole package feels like.

At the lowest setting, there is some gap between the spring and adjuster (or adjuster and subframe, since the adjuster isn't attached). It's not enough to worry about, and I'll never have it that low, anyway. I did drive it like that and it's just fine.

[BAT-man]

Ok good, now do fronts and then its MS3 taim!

[KevinK2]

Originally Posted by Code Monkey See here: Street car: 0.8 Hz Occasional autocrosser: 1-1.5 Hz Full-bore autocrosser: 2.2-2.5 Hz I bet those numbers were obtained by trial and error over a period of time and these days most everybody agrees these frequencies are a good starting point.

I have read that same link before, and it is an excellent guide. But, it's very important to understand that there are two different frequencies assosiated with spring selection. One is the spring-mass system of the unsprung weight. This is most important for street cars where you want the rear freq higher than the front by ~10%, so the impact response from both the front and rear wheels will be closer to being in phase.

But the other natural frequency, which that link addresses, is body motions/modes of the sprung weight. The main body natural frequency modes include the fore-aft pitching during hard braking and body lean during cornering. The method described in the link, and other racing articles, uses a simple approach to focus on the fore-aft pitching. The weight/mass used is the sprung weight at a corner, and the spring rate at the wheel.



This is another site, where Engineers who work(ed) in various technical trades answer questions. The suspension field is well represented by Gregg & Brian, who discuss natural frequencies in autocross:

www.eng-tips.com/suspension natural frequencies

Phate has done an excellent job on this thread, but this one issue of natural fequencies needs to be reviewed, imho.

.

[phate]

I've gone through the books posted above, (Puhn, Smith, Staniforth) and various articles by Ortiz and the folks at Optimum G, as well. I have yet to see a thorough explanation of how and why those ride frequencies are preferred. Even the "plan for worst bump" (1.5g, 2g bump, etc) is barely defensible by anything I've read. The "worst case" planning seems to result in extraordinarily high spring rates, which is where bump stops and shock valving come into play. The latter being the one thing I'm really relying on FCM for.

I'm not far enough along in my curriculum to offer an explanation, and I would like to find some text that offers it. If you have something to offer, or have some suggestions of reference material, I'd be glad to take a look or have a listen.

[KevinK2]

Originally Posted by phate I've gone through the books posted above, (Puhn, Smith, Staniforth) and various articles by Ortiz and the folks at Optimum G, as well. I have yet to see a thorough explanation of how and why those ride frequencies are preferred. Even the "plan for worst bump" (1.5g, 2g bump, etc) is barely defensible by anything I've read. The "worst case" planning seems to result in extraordinarily high spring rates, which is where bump stops and shock valving come into play. The latter being the one thing I'm really relying on FCM for. I'm not far enough along in my curriculum to offer an explanation, and I would like to find some text that offers it. If you have something to offer, or have some suggestions of reference material, I'd be glad to take a look or have a listen.

The main point I was presenting focused on the 2 different types of nf's that are involved in suspension design, as it seemed like you were persuing that approach, vs the 2g approach. I have read several analytical presentations of frequency based design. One is "New Directions in Suspension Design" by Colin Campbell, and in spite of the heavy math, he never came up with suggested frequencies to design to. You can find calculators for the wheel based nf, as well as the body/sprung wt type. Your Eibach link is based on the sprung wt "Sprung Weight: Vehicle corner weight less unsprung weight". This is the method my auto suspension engineers link supports.

I think the this is the best link, that includes a lot of your mods:

Autocross to Win (DGs Autocross Secrets) - Suspension

It provides natural frequencies to design to for autox, and it's based on the sprung wt. But you have to check the example he gives with mustang vs camaro to know that is his approach. THIS is WRONG --> My interpretation is that these nf's are higher than the speed related body mode frequencies that you would experience during an autox, as nf's are associated with amplifyed reactive motions.

EDIT: see post below for a proper explanation of natural frequency based spring design.


I also like this Ortiz article, as it is one of the few tha properly treats sway bars:

http://www.eviltwinmotorsports.com/w...ter-2011.2.pdf

I agree that the 1.5, 2g method can create overstiff suspensions. You have to include the bump springs and use an energy method that includes the dampers, vs a static analysis ... not easy.

Natual frequency analysis was one of the aspects of engineering I routinely practiced for industrial applications.


.

[KevinK2]

The best way to understand the sprung frequency target for luxury cars is to do a vertical body dance motion, with body rigid and bending at the knees in a smooth continuous motion. Then count the seconds to go from top positon, down and up again (1 cycle). If your time is 1 sec, it's quite smooth. That is 1 cycle/sec, or 1 hz.

If you go up-dn-up twice in one sec, that is 2 hz fequency, and represents what how you would be shaken in autox event. Get a girlfriend to try 3 Hz and see what happens

And based on this, I have a new appreciation of what the frequencies suggested by autox race shops:

It is the frequency that the car will try to bounce at, after a sudden G inertial load, like hitting the brakes, or getting a bit of lift off a rise in the pavement.

.

[KevinK2]

Originally Posted by phate Had some time to test fit the rear spring and adjuster setup, today. This is going to work perfectly. The only thing that sucks is the adjuster is a huge pita to adjust if there's any sort of pressure on it. When it's at the highest setting, you have to drop the LCA in order to relieve pressure. ....

I think that hastle is due to not having an adjustable, shorter travel shock, like in the coil-over kits you mentioned before. The B8 Bilstein will not be length adjustable, would a length and dampimg adjustable rear shock be available as a substitute? Hard to match the Bilstein quality.

.

[zenit]

First time I've had time to read this last few posts over.
Just to drop .2c in here...

about this

Originally Posted by Code Monkey See here: Street car: 0.8 Hz Occasional autocrosser: 1-1.5 Hz Full-bore autocrosser: 2.2-2.5 Hz

No doubt these are "t-shirt" sizes.
Some of these are generic fits for cars, with much more to be said about the fit.
Double A-Arm vs macpherson (camber compensation)
FWD vs RWD VS AWD
Weight & weight distribution
Roll Center & CoG
Anti/Pro Dive

I.E. MS3 can be tricky. With poor camber compensation in body roll, it makes sense to try to dial up the camber and roll resistance via spring rates. But with high spring rates comes a too fast response over smaller bumps and decreased acceleration potential. (I assume the WL AntiDive kit tries to address this issue. ) There is also sway bars to take into consideration, not only for roll couple but for turn in wheel rate to take into account.


In comparison, double A-Arm front suspensions have a better camber profile and afford softer springs and can afford to, or sometimes benefit from body roll. Last autocross, I remember one of the more senior members commenting about an ES miata - "they need to be tossed"

Another third item to keep in mind about frequencies is something that I've discovered about alignments. I have no doubt that they vary based on the available traction. A setup that works on a bumpy gravelly parking lot might be undersprung or conservative for concrete airport type surface.

In short: @phate, you're doing an awesome job.
I have no doubt you'll have a fantastic setup, but I'm sure there's more tuning in your future.

@KevinK2, IMO the criticism is appreciated, (I love bench racing suspension!), but I think phate's on the right track.

[KevinK2]

Sorry if I came across being critical. I was just trying to help out phate where he seemed to need it. Note I had to correct myself in red up there.

.

[phate]

Originally Posted by KevinK2 I think that hastle is due to not having an adjustable, shorter travel shock, like in the coil-over kits you mentioned before. The B8 Bilstein will not be length adjustable, would a length and dampimg adjustable rear shock be available as a substitute? Hard to match the Bilstein quality. .

I wasn't thinking when I was doing this stuff, apparently. I'm just going to unbolt the bottom of the shock next time to see if there's enough play in the bushings to drop the LCA low enough to unload the spring. Still doesn't change the fact that I have to jack up the car, but it would be a lot better than unbolting the LCA and sway bar every time.

I suppose an adjustable length shock (or adjustable mounting) would be nice, but like you said, it's hard to beat Bilsteins. I don't want to stray away from them, because of availability and the availability of revalving services. Having FCM investigate the MS6 for their Elite setups was probably the best thing to happen for MS6 suspension since the MS6 came out.

Preload on a spring doesn't change the rate, so our wheel rate and ride frequencies remain the same. Driving on the 600's between the no-load/preloaded states, I couldn't tell a difference in how the two felt. They were both slightly firmer than stock, but that's about it. There didn't seem to be much preload on the spring even at the tallest setting. It's just that the adjuster doesn't want to move with any load on it.

I originally wanted to go with some LCA weight jacks with spherical bearings that can be adjusted without jacking up the car. They are often used on BMW's, and could very easily be adapted to the MS6 LCA by cutting the stock perch and just welding in a flat plate. But, after reading the rules about 100 times trying to figure out a way to get around them, I came to the conclusion that it wouldn't be easily defendable in the case of a protest. Complete LCA replacements are allowed, but modification of the stock LCA's is not

Originally Posted by zenit I.E. MS3 can be tricky. With poor camber compensation in body roll, it makes sense to try to dial up the camber and roll resistance via spring rates. But with high spring rates comes a too fast response over smaller bumps and decreased acceleration potential. (I assume the WL AntiDive kit tries to address this issue. ) There is also sway bars to take into consideration, not only for roll couple but for turn in wheel rate to take into account. In comparison, double A-Arm front suspensions have a better camber profile and afford softer springs and can afford to, or sometimes benefit from body roll. Last autocross, I remember one of the more senior members commenting about an ES miata - "they need to be tossed" Another third item to keep in mind about frequencies is something that I've discovered about alignments. I have no doubt that they vary based on the available traction. A setup that works on a bumpy gravelly parking lot might be undersprung or conservative for concrete airport type surface.

Agreed. We did some testing on the FSAE car a couple months ago altering spring rates (it has no sway bars). I came in late to the game and they were working with ~3.2Hz front and ~2.9Hz rear frequencies. We went as high as 3.8Hz and as high as 3.3Hz in the rear. Going up in rate on either end reduced grip significantly and turned the car into a turd. That was a real eye opener seeing what too much spring really does.

Double A-Arm production cars do gain camber in bump travel, which is a beautiful thing. [Mapping camber curves is what started this whole investigation and then building my own coilovers, lol.] But, it's typically not enough to compensate for body roll, so you still need to control roll to maximize grip. Miata's have some pretty extreme camber gain - Something like 1.1° per inch of wheel travel on the NA's (even more in the rear, I believe), whereas the MS6 gains ~.75° per inch of wheel travel.

Originally Posted by KevinK2 Sorry if I came across being critical. I was just trying to help out phate where he seemed to need it. Note I had to correct myself in red up there.

Not at all, I appreciate the input, and I think anyone else reading this thread will, too. Like I said, I'm learning things as I go. I've never gotten this deep into suspension before, so I'm taking my usual approach: obsess and learn, sift through the bs, try to apply, fine tune, repeat.

[KevinK2]

Originally Posted by phate .... I suppose an adjustable length shock (or adjustable mounting) would be nice, but like you said, it's hard to beat Bilsteins. I don't want to stray away from them, because of availability and the availability of revalving services. Having FCM investigate the MS6 for their Elite setups was probably the best thing to happen for MS6 suspension since the MS6 came out.

They are top notch. The Spitfire datalogging video showing vibration in in/sec, displacement, car speed, and displ at the shocks and G's inside the car, was just like what I would be looking at troubleshooting industrial machines.

Preload on a spring doesn't change the rate, so our wheel rate and ride frequencies remain the same. Driving on the 600's between the no-load/preloaded states, I couldn't tell a difference in how the two felt. They were both slightly firmer than stock, but that's about it. There didn't seem to be much preload on the spring even at the tallest setting. It's just that the adjuster doesn't want to move with any load on it.

It sounds like at full rear droop based on shock travel, the rear spring was not loose, but had some slight preload at both the high setting and the low setting you tried. This may be due to an internal bushing in the shock to avoid a hard hit when you experience full travel while driving.

For your custom design (like early Ground Control), both front and rear, it seems like your preload setting (if any) is also the ride height setting. When you adjust the front/rear spring "preload", you are actually setting ride height too. For example true front coilovers have the spring supported by a double nut, that sets the preload. The height is set by how the external threaded tube screws into a lower housing, with a single jam nut, so you can change ride height without changing preload. Sorry if you already covered this.

Your F/R Springs

.... We did some testing on the FSAE car a couple months ago altering spring rates (it has no sway bars). I came in late to the game and they were working with ~3.2Hz front and ~2.9Hz rear frequencies. We went as high as 3.8Hz and as high as 3.3Hz in the rear. Going up in rate on either end reduced grip significantly and turned the car into a turd. That was a real eye opener seeing what too much spring really does.

FSAE is excellent training/education. But for that small a vehicle, was the unsprung natural frequency analized, or the sprung nf? I would guess the unsprung.


.

[phate]

Originally Posted by KevinK2 They are top notch. The Spitfire datalogging video showing vibration in in/sec, displacement, car speed, and displ at the shocks and G's inside the car, was just like what I would be looking at troubleshooting industrial machines. It sounds like at full rear droop based on shock travel, the rear spring was not loose, but had some slight preload at both the high setting and the low setting you tried. This may be due to an internal bushing in the shock to avoid a hard hit when you experience full travel while driving. For your custom design (like early Ground Control), both front and rear, it seems like your preload setting (if any) is also the ride height setting. When you adjust the front/rear spring "preload", you are actually setting ride height too. For example true front coilovers have the spring supported by a double nut, that sets the preload. The height is set by how the external threaded tube screws into a lower housing, with a single jam nut, so you can change ride height without changing preload. Sorry if you already covered this. Your F/R Springs.

Correct. The only preload I have is the load from the chassis to the spring. There's no way to change the preload without changing ride height.

It's nearly the same with the rear. Since the spring/adjuster is longer than the space available, there is a small amount of preload when the ride height is set high (I'd guess ~half of the adjustment range causes preload). When the adjuster is set to the lower half of it's adjustment range (lower ride height), there is no preload at full droop. It's a very small amount - not enough to let the spring come misaligned if the suspension were to see full droop during operation.

This seems acceptable for most door cars, since the ARB will lift a wheel and keep the spring seated even if we were to transfer all the weight from a corner. [Not that I want to lift a wheel, it's just an example.] Adjustable length shocks like you are talking about seem very uncommon for most applications.

Side note: The FSAE car does transfer all weight off of the front corners, which was really incredible to see. Like I said, I came into the program after the car was designed, built, and running; so the suspension adjustment has been wonky and we've been sorting it out. This video shows the spring coming unseated in a long sweeper at autox.

Originally Posted by KevinK2 FSAE is excellent training/education. But for that small a vehicle, was the unsprung natural frequency analized, or the sprung nf? I would guess the unsprung. .

We used Staniforth's approach to analyze the sprung NF. We have the telemetry to actually log the bounce frequency, which we need to do at some point. But, we have testing time available and we have the telemetry to dial things in based on it, so we're running with it.

I don't think I've ever seen an analysis of unsprung frequencies in my reading and searching.

[KevinK2]

Originally Posted by phate Side note: The FSAE car does transfer all weight off of the front corners, which was really incredible to see. Like I said, I came into the program after the car was designed, built, and running; so the suspension adjustment has been wonky and we've been sorting it out. This video shows the spring coming unseated in a long sweeper at autox.

You mentioned before how a sway will prevent full droop in your case. In my case, with big bars bolted to a 924Turbo, I have a picture of the right front tire comming off the ground in an 80mph right hand sweeper! Bar was a bit too big.

I don't think I've ever seen an analysis of unsprung frequencies in my reading and searching.

I read it once in that "heavy math" book I mentioned, and saw it in one of those calculators online. I've seen cars on the road with bad shocks, and a tire bouncing frantically. If you watch Nascar on the road course in Calif, sometimes as they come down a high speed hill and that big fancy solid axle in the rear gets bouncing under heavy braking. That would be at the rear unsprung weight nf.

But you are correct, the NF's here are correctly based on sprung wt. At first I thought you were considering unsprung only, but I was wwwrrrong.

.

[phate]

I test fit the front spring perches this past weekend and they are pretty much perfect There is a TON of room above the shock tower in an MS6 - these plates could be raised by another 1+ inches before hitting the hood.

I trimmed the spring perches down so they are flat on the shock tower side.



This just shows how well the holes line up. The shock should be able to pass right through the shock tower if needed.





Here's the assembly with shock mounting plate and bushings:


^I just picked up a set of Energy suspension shock bushings for like $10. They are part # 9.8126.

Rather than relying on just that bolt to hold up the shock mounting plate, I made some solid threaded adapters for it to sit on. Should help keep the assembly more rigid. I made some .5" and 1" tall spacers to have some adjustment. If I want to raise the shock mount further, I'll make it out of steel and use a bolt on the top and bottom, rather than just one long bolt with a nut on top. I don't think there's a need to raise it further, based on bump travel limitations of the upper control arm, though.

[phate]

Finally made some time to measure the front suspension pickup points and whatnot. Here's what it looks like:


^The vertical dashed lines represent the hub mounting surface. The center of the circles represent the center of the tire contact patch of the stock wheels/tires.

Here's what the instant centers and roll center look like at stock ride height:

^roll center is 7.15" below LCA pickups.

Here's what the instant centers and roll center look like when you have the car lowered to near H&R height (just for reference):

^Roll center is 6.98" below LCA pickups.

You might say, "phate! how do you know what the control arms are doing during bump travel"? Well, Johanna and I took measurements of them a while back. Angles of the control arms through bump travel:

^90° represents horizontal, <90° represents the outer pickup (ball joint) lower than the inner pick, >90° represents the outer pickup higher than the inner pickup. The horizontal axis represents hub center to fender measurement, starting at full droop.



I'll get the rear suspension layout some other time. Once we have that, we'll have a better idea of roll axis (I'm just working with some assumptions right now).

I built a spreadsheet, based on Allan Staniforth's book, that calculates front/rear roll resistances, frequencies, and roll angles based on corner weights,center of gravity height, roll centers, and spring rates. It was originally for the FSAE car, so I don't have roll bars included, which is probably my next step with it. I'm going to expand it to include them hopefully soon. We'll need to make some educated guesses about the ARB rates, but it should get us pretty close.

I'll add the sketches to the OP to keep it all consolidated in one place.

[KevinK2]

Nice work with solidworks ... do you have an analysis package too?

Just got back from vacation and doing taxes .... I'll get back to comparing your dimensions with my M6 ones in my Puhn SS. Then we can compare results for a 1G corner on your lowered MS6, and the oem version. There is also an online suspension program that I ran across on this site: farnorthracing_autox_susp calculator It would be great to cross check all three programs.

About the rear ASB rates, I derived Puhn's published formula and it is valid for bars that are straight between bushings, like the front bar. But for that rear bar, I was thinking of a bench test. That is what is the basis of Puhn's formula ... one end pinned, load at other end. I have a very stiff work bench that I had started testing the Eibach tubular bars on the Rx7. But as I started on the rear bar and increased load, the bar was seeing too much deflection. Turns out I had cracked the strait length of the bar at the last track day (though it was me when the car started to push in the turns).

Another way to test the bar is in place on the car. You would need to fix one end relative to the frame (could be from garage floor if frame is on jack stands) , and then apply loads to free end and use a dial indicator for deflection.

Also need to find out the height of the Sprung Weight CG. Testing with scales is a lot of work .... some say engine cg is a good estimate ...

Cool to do this right !



.

[phate]

The version of SolidWorks I have on the laptop is pretty bare bones. I'll see if I have the motion analysis stuff available here. I bet we do. I'm not sure what online program you're referring to, but I came across this one with some quick googling. Haven't had time to use it other than dragging components around, though.

If nothing else, I could just build some solid links (the quadrilaterals that make up the control arms) and put hinge joints at each corner and fix the inside link. Can you make a driven dimension off of that outside link? I haven't worked with SW much recently; I just have it on my laptop. I've been learning Creo most recently. Maybe I'll just whip that up tonight.

I called Whiteline about the bar's actual rate and, not surprisingly, they don't test them for exact rates. A test rig would be pretty cool (and simple), but I doubt I could do that any time soon (and I don't want to pull the bar). Testing it in car with the LCA's hanging out of the way would be super easy, though...I'll have to think about this a bit more, but I can probably come up with something. Biggest issue would be getting enough weight hanging. [Could also do this for the front bar very easily.]

I'm just going to use the engine centerline as the CG estimate. Finding the Z-axis CG is a pita, even on a tiny formula car.

[KevinK2]

I edited my post to include the free online link. The one you found goes beyond the scope of what I was considering. The CAD type programs I have used were included in my FEA programs, NASTRAN and ABAQUS ... industrial priced stuff.

For my ARB test bench, I used some lifting weights I have including several 25 lb metal weights, with aircraft cable through the hole. I had about 50 lbs on the bar that had broken on the track. I'll list some of the Mazdaspeed6 bar rates as a reference later.

.

[KevinK2]

I ran some spring rates for the available Front sway bars. I measured my bar's geometry, which is the same as the MS6 (check aftermarket bars, usually spec'd for both cars).

Puhn uses these dimensions, centerlines to hole centers:

A -- 8.25 perpendicular length of arm
B -- 22.0 straight span
C -- 11.0 length along arm

Assumes these dimensions are the same for each bar.

Dia / Bench stiffness / wheel rate (constant speed corner) lb/in

24.0 / 413 / 326 (MR = .63, wheel rate = 2X bench rate)
25.0 / 486 / 384
26.0 / 569 / 450
28.5 / 822 / 648

.

[phate]

Thanks for those. I'll incorporate ARB's into my spreadsheet probably next week.

I've added info about Whiteline bushings to the modification sections. I added sway bar end link dimensions to the OP. I'll have more info about making end links later on. Waiting to hear back from a company, so we might have a cheaper/better alternative to AWR's rod ended links.

[KevinK2]

Puhn's beam based sway bar formula:

Automotive suspension engineering - Sway bar spring rate formula

This is bench test stiffness. When doing an analysis of a ss corner, enter 2X this stiffness, and use MR^2 to get stiffness at tire.

.

[KevinK2]

These are the dimensions I measured for my M6, how do they compare with your graphic results?


Front Suspension - Lever Arms, inches
rgt to lft pivot spread --- 28.75
to sway bar --- 6
to spring --- 9.3
to b-jt --- 13.5
b-jt to wheel ctr --- 3.125

Rear Suspension - Lever Arms
rgt to lft pivot spread --- 15
to sway bar --- 5.5
to spring --- 10
to b-jt --- 18
b-jt to wheel ctr --- 5.5

----------------------------------------------
This is data for a stock M6:


Height, CG Sprung Weight --- 17 est'd
Height, CG unsprung weight --- 12.4 est'd
Height, Front Roll Center --- 6 (graphically determined from illustration)
Height, Rear Roll Center --- 8 est'd
Height, Roll Center @ CG --- 6.8

[KevinK2]

Originally Posted by KevinK2 These are the dimensions I measured for my M6, how do they compare with your graphic results?

Lets see:

Front Suspension - Lever Arms, inches
rgt to lft pivot spread --- 28.75 (29.75)
to sway bar --- 6
to spring --- 9.3 (9.05)
to b-jt --- 13.5 (12.43)
b-jt to wheel ctr --- 3.125 (3.36)

Your drop string measurments are more accurate than my tape measurer. For the last dimension, I started with the specified track width. It looks like you are using 48mm offset wheels.

Rear Suspension - Lever Arms
rgt to lft pivot spread --- 15
to sway bar --- 5.5
to spring --- 10
to b-jt --- 18
b-jt to wheel ctr --- 5.5

----------------------------------------------
This is data for a stock M6:


Height, CG Sprung Weight --- 17 est'd
Height, CG unsprung weight --- 12.4 est'd
Height, Front Roll Center --- 6 (graphically determined from illustration) (2.48)
Height, Rear Roll Center --- 8 est'd
Height, Roll Center @ CG --- 6.8

Your front roll center from Solidworks is more accurate than my use of a Mazda Service Highlights illustration, extended with extra sheets of paper ...

Regarding the debate about the front motion ratio, to use anything but the MR from your accurately measured lever arms (.75 vs .57) is likely just making up a fudge factor to cover up an error in Fat Cat's program ... note .75^2 = .56 ! Note that even with my crude measurments, I had a rear MR of .56 . This is for a M6, but I suspect the parts are similar in spite of the extra half-shaft.

.

[Tomas]

Since you can get individual components from BC. Why don't you get the upper spring perches for the front from them as well? Pretty sure they use a 2.5 ID spring.

[phate]

Probably could and would work fine. I was originally going a different route and had the perches made, then changed my mind and went with the current design.

[cnyms6]

@phate; do you have any news on your FCM damper rebuilds? I'm excited to hear how this suspension performs!

[phate]

They're "almost" finished. The original due date was March 29th, and it's been pushed back numerous times. I'm not happy about the whole situation. One of the reasons I did this was because of the lead time on a full kit from them. A revalve was supposed to be much faster...

I can't really recommend FCM unless you have lots of time to wait.

[Deldran]

Who else is there to go with

Sent from my gs5

[phate]

Bilstein USA will do a revalve based on whatever info you supply. There is some question about how good they are, but it's also less than half the cost of FCM...

[phate]

I received word yesterday that my shocks were being taken apart for the revalve, finally. Might have them back next week and maybe we'll see this project come to fruition. My next autocross weekend is July 19th and 20th. Hopefully it'll all be together by then.

I'm going to need to make some bump stop shims so I can adjust the contact point up front. For right now, no bump stops will be used in the rear. We'll see how it feels without them, and maybe make some sort of adapter back there to mount them if necessary. I'm not planning on lowering the car much, so not having them isn't really a risk.

[KevinK2]

when did you request the re-valving?

.

[phate]

~4 months ago.

[sidekick]

After reading this thread and watching some of Shaiks videos, I really want to have him build a coilover setup. A smooth-ish ride with great handling sounds amazing, but I also don't want to wait 2 years for suspension. Lol. This thread is awesome btw, I learned a lot and I still have a lot more to learn when it comes to suspension.

[Code Monkey]

Originally Posted by sidekick After reading this thread and watching some of Shaiks videos, I really want to have him build a coilover setup. A smooth-ish ride with great handling sounds amazing, but I also don't want to wait 2 years for suspension. Lol. This thread is awesome btw, I learned a lot and I still have a lot more to learn when it comes to suspension.

The "2 years" included all the R&D so don't be discouraged by my experience. Even when Shaikh has already perfected a particular setup, as he did for Miatas, the wait time is notorious. Search any Miata forum and see that a 4-week delivery time always turns out into 12+ weeks. I would say 9 out of 10 users say they would do it again, including the wait, while the remaining 1 user says he would never ever do business with Shaikh again. From what I heard, to motivate Shaikh to deliver faster, it helps to pay 50% of the invoice and insist that you will pay the remaining 50% once the coilovers are ready to ship.

[Mazdazilla6]

So I've read this thread for the third time now and I'm happy to say I understand a good chunk of it. The big cloud still in the sky for me is how sway bar rates tie together with spring rates. PS if you need a stock rear bar to test @phate; I have mine lying around.

Also I've been tossing around the idea in my head about a good coilover option for us and when the time comes for me to buy a somewhat good suspension I will get BC's and get swift springs on the front (which is an option through BC) and hopefully there will be another option for rear springs to work with BC perches other than QA1 as I think those would be too stiff for me (I don't plan on auto-x that much so I think 1.50-1.60 Hz would be sufficient) and the BC rear springs are soft poo. As for shocks I think the BC ones are fine for an occasional auto-x and having on the spot adjustability is always a plus even if they don't compare to Bilsteins.

Thanks again for the good read Clint!

[Code Monkey]

@Mazdazilla6; when I was ordering my BC Racing perches I talked to the guy on the phone quite a bit and he said they did custom spring rates and revalves and that I could have ordered a coilover kit for a Mazdaspeed6 from them with 1000/600 spring rates. I do not know how truthful those statements were but call them and talk.

I also emailed Swift separately inquiring about tapered 6" springs for the rear, they said they did not make tapered springs, but advised me to contact BC Racing who "should have a bracket that allows you to use a linear coilover spring in the stock location. Its usually a bracket that fits up top and a threaded perch on the other end. You will also need to ask them what length spring they recommend with that setup." @phate; if this bracket exists would running a 2.5" ID spring top and bottom in the rear work for us?

[phate]

Originally Posted by Mazdazilla6 So I've read this thread for the third time now and I'm happy to say I understand a good chunk of it. The big cloud still in the sky for me is how sway bar rates tie together with spring rates. PS if you need a stock rear bar to test phate I have mine lying around. Also I've been tossing around the idea in my head about a good coilover option for us and when the time comes for me to buy a somewhat good suspension I will get BC's and get swift springs on the front (which is an option through BC) and hopefully there will be another option for rear springs to work with BC perches other than QA1 as I think those would be too stiff for me (I don't plan on auto-x that much so I think 1.50-1.60 Hz would be sufficient) and the BC rear springs are soft poo. As for shocks I think the BC ones are fine for an occasional auto-x and having on the spot adjustability is always a plus even if they don't compare to Bilsteins. Thanks again for the good read Clint!

Sway bars simply add additional roll resistance on that end of the car without altering sprung mass frequency at normal ride height. [They add roll resistance in roll, lol.] Remember that weight transfer doesn't change with additional roll stiffness - only the perceived affect of weight transfer changes, the roll of the sprung mass (body roll).

You can get the same amount of roll stiffness from either springs or bars, or a combination thereof. If you are shooting for a specific total roll resistance, and are doing it with only springs, you run the risk of having exorbitantly high frequencies. So, for me, I put a higher priority on sprung mass frequency, and I'll tune total roll resistance and the proportion of roll resistance with ARB's. When I talked to Shaikh about suspension stuffs, he said that his priority is 1) sprung mass frequency, and 2) flat ride, and 3) final roll control with ARB's. [That was kinda cool hearing that I'm on the same track as him.]

As for your suspension, it comes down to what you're comfortable with. I expect mine to be firm, but not at all harsh. Come down to an autocross some time and you can ride in it and I'll let you take it for a spin around town.

Originally Posted by Code Monkey Mazdazilla6 when I was ordering my BC Racing perches I talked to the guy on the phone quite a bit and he said they did custom spring rates and revalves and that I could have ordered a coilover kit for a Mazdaspeed6 from them with 1000/600 spring rates. I do not know how truthful those statements were but call them and talk. I also emailed Swift separately inquiring about tapered 6" springs for the rear, they said they did not make tapered springs, but advised me to contact BC Racing who "should have a bracket that allows you to use a linear coilover spring in the stock location. Its usually a bracket that fits up top and a threaded perch on the other end. You will also need to ask them what length spring they recommend with that setup." phate if this bracket exists would running a 2.5" ID spring top and bottom in the rear work for us?

Yes, it's sorta true that BC will do custom rates and custom valving for you...I talked to them about it a long time ago. You know what they quoted me for a dedicated autocross setup? An OTS kit with swift springs. I then asked them about what the sprung mass frequencies would be with their setup, and you know what? They had no idea. They said, "we don't research individual chassis'"...How in the world do they valve for that, then? How did they come up with those spring rates? What's the proportion of roll resistance going to be? They give you KNOBS, and hope you'll be happy turning them even though you don't know what the heck they do. Do they increase stiffness the same amount each click, do they have cross talk, do they adjust to the same damping if you come at it from a different direction, blah, blah.

They really pushed for knobs (DIAL IT IN BRO), and I kindly sent them a huge email explaining that I didn't have the telemetry to dial in something that they couldn't even come up with a suggestion for. I wasn't real happy with them, and that was just another push towards building my own. Anyway, I digress.



For the 2.5" spring in the rear - not unless you want to cut out the bump stop mount. That was my original intent, actually, because I wanted a standard ID race spring. The bottom of it is flared to some diameter that is >2.5" (I can't find my drawing right this second).